A Critique of Animal Psychology Research at the University
of California at Berkeley
Brandon P. Reines, D.V.M.
Mental imagery research: Dr. Walter J. Freeman
Dr. Freeman has received $1,718,889 from the NIMH since 1963. Freeman's primary research involves removing one eye of a rabbit. Then, electrodes are threaded through the back of the eye socket into the olfactory bulb, a part of the brain that receives information about odor. When the rabbit smells food, a specific pattern of electrical activity appears, which Freeman claims is caused by the rabbit's "mental image" of food. Freeman uses approximately 12 rabbits per year for his research. Freeman has been conducting physiological psychology experiments on animals since the late 1950's. In an effort to justify his psychological experiments on animals, Freeman recently wrote:
...my research has led to three main advances. One is in demonstrating by physiological means that the brain is not a computer, and to give a strong scientific basis and encouragement for the movement among psychiatrists and psychologists for the existence and potential therapeutic utility of mental images ... Another is a new set of techniques which (are) not yet in clinical use, will provide clinicians with increased power for use of EEG and MEG in the diagnosis and treatment of functional disorders of the brain ... The third is a model for epilepsy of a common type among children, which I have reproduced in cats, rats, and rabbits, and which I have recently demonstrated mathematically is due to runaway inhibition rather than runaway excitation as commonly supposed. This process had been suspected, but had not been accepted, being counterintuitive, so that treatment of epileptic children with GABA and related inhibitory substances have been pursued despite lack of benefits ... my animal research will continue to support clinical advances in the future. The basic fact is, we do not yet know how the brain works. Therefore, we cannot rationally know how it goes wrong and what to do about it for a host of diseases ranging from hypertension and cardiac death to suicidal depression and the schizophrenias. The answer will not be found in tissue culture, because cells there do not organize themselves into brains. Nor will it be found in computer simulations; we use them heavily for proofs, but we must know what we are simulating. That can only come from animal studies, because it is inadmissable to place electrodes into the brains of humans other than for therapeutic purposes. I repeat my firm belief: The only path to understanding the brain and its disorders is through animal research (1986b).
A point-by-point critique of Freeman's rationale follows:
1) Freeman's first claim was simply a linguistic distortion. It was well known that the brain is not a computer long before the advent of science. Freeman is really saying that the brain-as-computer analogy does not seem to be providing insight into human brain function. In other words, no one really knows how the human brain "works"; ignorance is not an advance, however. Freeman (1983) further maintained that his animal research on "the physiological basis of mental images" has led to treatments that utilize mental imagery. In fact, psychologists were using therapies that employ the concept of mental imagery long before Freeman's work. New therapies such as "creative visualization" are coming into vogue for a variety of reasons that have nothing whatever to do with animal research.
Just what psychologists mean by "mental image" is in fact an open question. One critic of the field of mental imagery wrote:
Any analysis in the nature and role of imagery is fraught with difficulty. The concept itself is difficult to pin down. Is a visual image like some conceivable picture? If not, then in what way must it differ? ... The tumultuous history of the concept of imagery in both philosophy and psychology attests to the difficulties which such questions have raised in the minds of scholars for centuries. While contemporary psychologists have attempted to narrow the scope of the concept by operational definitions and multiple empirical and theoretical underpinnings, it is not clear that they have resolved the major conceptual ambiguities, circularities, and Rylean 'category mistakes' which have plagued the notion in the past (Pylyshyn, 1973).
If the notion of a "mental image" is unclear in human beings, it is even less clear in laboratory animals. Freeman himself concedes that he is on "shaky ground" in referring to mental images in animals:
If this line of reasoning is correct, then these EEG contour plots reveal our first glimpse of the physical aspect of a mental event. Even though we cannot know whether rabbits introspect, I have elected here to call it a 'mental image' partly because I believe that neural and mental images are two sides of the same coin, and partly because I want to attract the attention of cognitive psychologists, who believe that such 'images' or 'representations' must exist, and who now vigorously debate their nature ... (1983, p.1121).
Neither of Freeman's two reasons for referring to a "mental image" in the rabbit psyche are based on actual evidence. Freeman's belief that the rabbit experiences a mental image is not evidence for such an image; surely, using the words "mental image" to attract the interest of cognitive psychologists is not a scientific justification for the concept.
2) The second claim was so loosely worded that it is difficult to interpret. Freeman asserted that his research will somehow increase the "power" of clinicians. Rather than claiming to have achieved a bona fide advance, it would have been more honest for Freeman to say that he believes that his research may one day prove useful to clinicians. Such a statement would reduce his claim to what it is - pure speculation. There is really no scientific reason to believe that his research will lead to improvements in the treatment and prevention of any brain disorder.
3) Freeman's third claim was that he has created an animal model of a type of epilepsy in children. Aside from the question of whether the existence of an animal model in and of itself constitutes an advance, there is no evidence that he has created such a "model." Freeman's only apparent attempt to validate his "model" by comparing it to human seizures is indicated by the following quote:
The unpredictability in the detail of the simulated and recorded seizure spike trains indicates that they are chaotic. The same conclusion was reached by Babloyantz and Destexhe from dimensional analysis of scalp EEG recordings during human epileptic seizures" (1986a, p. 20).
The goal of science, of course, is to find order in the apparent chaos. It is hardly an advance to reproduce chaos. Freeman himself conceded: "This model is preliminary and incomplete." (p.21).
In the second part of Freeman's argument, he maintained that in order to treat hypertension, cardiac death, depression, and schizophrenia, it is necessary to understand how the brain works. There have been substantial advances in the treatment of all of these disorders, such as chlorpromazine for schizophrenia, without "understanding how the brain works."
1. Hormones and Behaviour Research
3. Mother-Infant Separation Research
5. Mental Imagery Research